What is hapai CLAUDE.md?

Repository instructions for hapai, a Bash-based guardrails system for AI coding assistants. It uses hooks to inspect tool calls and block rule violations before they run.

Use it when working on hapai's hooks, Svelte dashboard, BigQuery or Cloud Storage integration, platform exporters, or development environment.

It explains what hapai is, what it requires, and how its dashboard, cloud integrations, and exporters fit together. This reduces setup mistakes when developing the repository.

hapai CLAUDE.md — Instructions file for coding agents by renatobardi: Repository instructions for hapai, a Bash-based guardrails system for…

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add instructions/renatobardi/hapai/claude-md

Clone the repo

git clone --depth 1 https://github.com/renatobardi/hapai

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/renatobardi/hapai/claude-md.svg)](https://agentmods.dev/instructions/renatobardi/hapai/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/renatobardi/hapai/claude-md"><img src="https://agentmods.dev/badge/instructions/renatobardi/hapai/claude-md.svg" alt="Measured on agentmods" height="20"></a>

Per session 9,606 This file is loaded in full into every session.

When invoked 9,606 The same file — it is already loaded in full.

Security scan F 4 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.09606	$0.09606
Opus 5	$0.04803	$0.04803
Sonnet 5	$0.01921	$0.01921
Haiku 4.5	$0.00961	$0.00961

Measured 5d ago against content hash 10bc3d259acf, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade F, and why

hapai CLAUDE.md scanned grade F with 4 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Downloads and executes remote codehighSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

curl -fsSL https://raw.githubusercontent.com/renatobardi/hapai/v1.8.0/install.sh | bash

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

- `HAPAI_DEV=1` installs hooks to `~/.hapai/hooks/` and registers them in `~/.claude/settings.json`

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

echo '{"hook_event_name":"PreToolUse","tool_name":"Bash","tool_input":{"command":"rm -rf /"}}' | \

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

├── install.sh # Universal installer (curl-safe)

CLAUDE.md · 887 lines

How it starts

The opening of the file, as written. The whole thing — 887 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

What is hapai

hapai is a deterministic guardrails system for AI coding assistants (Claude Code, Cursor, Copilot). It enforces security rules via shell-based hooks that intercept tool calls and block violations before execution — not probabilistic prompts that can be ignored. Pure Bash, only external dependency is jq.

The system combines:

Hook enforcement — Shell scripts that run before/after tool execution
Svelte 5 Analytics Dashboard — Real-time guardrail event visualization
Cloud integration — BigQuery + Cloud Storage + GitHub Pages deployment
Multi-tool exporters — Export guardrails to Cursor, Copilot, Windsurf, etc.

Prerequisites

Before working on this codebase, ensure:

jq 1.6+ — Required for all hook scripts and validation. Install: brew install jq
Node.js 20+ + npm — Required only for dashboard development (infra/gcp/dashboard/)
Bash 4+ — All scripts use set -euo pipefail and POSIX-compatible tools
Git — Version control and release tagging
Tested platforms — macOS and Linux. CI runs on both via ci.yml. WSL supported via installer detection.

Development Environment Setup

Complete setup for contributing:

# 1. Clone the repository
git clone https://github.com/renatobardi/hapai.git
cd hapai

# 2. Verify prerequisites
jq --version      # Should be 1.6+
bash --version    # Should be 4+
node --version    # Should be 20+ (only if developing dashboard)

# 3. Install for development with hooks enabled
HAPAI_DEV=1 bash install.sh

# 4. Validate installation
hapai validate

# 5. Run full test suite to verify everything works
bash tests/run-tests.sh

# 6. For dashboard development, install Node dependencies
cd infra/gcp/dashboard && npm ci && cd ../..

# You're ready to develop!

Development setup notes:

HAPAI_DEV=1 installs hooks to ~/.hapai/hooks/ and registers them in ~/.claude/settings.json
Project-local hapai.yaml overrides user-global ~/.hapai/hapai.yaml — useful for testing
All hook changes take effect immediately; no reinstall needed
Dashboard requires .env file with Firebase credentials (see Dashboard section)

Read the full file on GitHub · 887 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

5d ago First seen · 887 lines · 9,606 tokens per session scan F 10bc3d259acf

Subscribe to this mod's changes

hapai CLAUDE.md is an instructions file published in the GitHub repository renatobardi/hapai (16 stars, last pushed 4mo ago), licensed MIT. It adds 9,606 tokens to every session, about $0.0480 per session on Opus 5. A static security scan graded it F with 4 findings (downloads and executes remote code, reads agent configuration directories, recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

AI Safety Guardrails Security bash bigquery cloud-functions svelte

hapai CLAUDE.md

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade F, and why

Downloads and executes remote codehighSupply chain

Reads agent configuration directoriesmediumAgent snooping

Recursive force deletehighDestructive command

Makes network callslowCapability

How it starts

CLAUDE.md

What is hapai

Prerequisites

Development Environment Setup

What this file has done since we first saw it

Other instructions, from other repositories

Doberman-Core AGENTS.md

ThumbGate CLAUDE.md

ThumbGate AGENTS.md

lanekeep CLAUDE.md

FeatherlaneAI AGENTS.md

ThumbGate GEMINI.md